The first time I visited GitHub's HQ2 in 2012, they had a TV showing off their first animations of Mona and using it to push their new tagline: Social Coding.…

Justin Searls

AI coding agents are driving a shift toward what the author calls 'Antisocial Coding' — a trend away from collaborative, social development practices. Three key observations: agents create hub-and-spoke communication bottlenecks that make multi-developer collaboration unwieldy; agent-built codebases calcify quickly into overly complex, single-owner projects; and open source projects are closing off contributions (e.g., tldraw removing its test suite to prevent agent-assisted rewrites). The downstream effects include increased bus-factor risk, erosion of collective code ownership, and fewer mentorship opportunities for junior developers. Engineering leaders are urged to consider these organizational impacts before optimizing purely for agent productivity.

Agents are ushering in the Antisocial Coding era